As the manufacturing paradigm shifts from Industry 4.0 to Industry 5.0, industrial systems are transitioning from system-centric to human-centric approaches. Industrial assembly remains a domain highly dependent on skilled manual labour, posing challenges to automation and collaborative efficiency. While enabling technologies such as Human-Robot Collaboration (HRC) and Digital Twins (DT) hold promise, current approaches to modelling human behaviour, robot behaviour, and task structure remain fragmented and lack integration. This disconnect limits the effectiveness, safety, and adaptability of HRC in complex industrial settings. To bridge this gap, this paper proposes a structured hierarchical model that integrates task, human, and robot behaviours within an industrial HRC context. Using Design Science Methodology (DSM), the study constructs a structured framework grounded in Hierarchical Task Networks (HTN) and computational ethology principles. The model decomposes tasks into multiple abstraction levels: primitive tasks, atomic actions, motion primitives (MP), and motion frames. Experimental demonstrations show how this hierarchical structure supports motion clustering, behavioural prediction, and safety control. The proposed model enhances transparency, adaptability, and coordination, advancing the implementation of safe and efficient HRC aligned with the goals of Industry 5.0.
Xia et al. (Wed,) studied this question.